Skip to main content
Version: 14.3.0

Building a REST API on Top of a Legacy Database

The Scenario​

A retail company has a legacy MongoDB database that contains all of its product inventory information. A new mobile application needs to access this data to show product availability to customers. The development team needs to create a modern, secure, and scalable REST API to expose this data, but they don't have the time or resources to write a dedicated microservice from scratch.

The Challenge​

  • Time Constraints: The project has a very tight deadline, and writing, testing, and deploying a new API layer would take several weeks.
  • Lack of Expertise: The team primarily consists of frontend developers and lacks deep expertise in backend development and API design best practices.
  • Need for Standard API Features: The API must support standard features like pagination, filtering, sorting, and projection to be efficient for the mobile app. Implementing these from scratch is complex.
  • Security Requirements: The data is sensitive, so the API must be secured, and access must be controlled.

The Solution with Mia-Platform​

The team decides to use Mia-Platform Console to create a secure and fully functional REST API without writing a single line of code.

  1. Connecting the Database: First, the team configures the connection to the legacy MongoDB database. They create a secret variable in the project's environment settings containing the MongoDB connection string. This ensures that the credentials are kept secure.

  2. Creating the CRUD Service: From the Marketplace, they add the CRUD Service plugin to their project. In the service's environment variables, they reference the MongoDB connection string variable they just created. The CRUD Service is a ready-to-use component that automatically handles the logic for interacting with a MongoDB database.

  3. Defining the Data Model (CRUD Collection): In the MongoDB CRUD section of the Design Area, the team defines a new "CRUD Collection" named products.

    • Instead of manually defining all the fields, they use the "Import fields from data sample" feature. They provide a small JSON file with an example of a product document from their legacy database.
    • The Console automatically infers the data schema, including field names and types (string, number, boolean, etc.). The team quickly reviews and confirms the schema.

  4. Exposing the REST API (Endpoint): In the Endpoints section, they create a new endpoint:

    • Base Path: /products
    • Type: CRUD
    • CRUD Base Path: They select the /products internal endpoint exposed by the CRUD Service.

  5. Securing the Endpoint: In the endpoint's security settings, they check the "Authentication required" box and define a User Group Permission (e.g., groups.mobile_app) to ensure that only authenticated clients belonging to the correct group can access the API.

  6. Deploying the Solution: The team saves the configuration and deploys the project.

The Outcome​

  • API Ready in Minutes: In less than an hour, the team had a fully functional, production-ready REST API exposing their legacy data. There was no need for backend coding, which would have taken weeks.
  • Rich Features Out-of-the-Box: The API automatically supports:
    • GET /products/: To list all products, with support for pagination (_l, _sk), filtering (_q), sorting (_s), and projection (_p).
    • GET /products/:id: To retrieve a single product.
    • POST /products/: To create a new product.
    • PATCH /products/:id: To update a product.
    • DELETE /products/:id: To delete a product.
  • Secure by Design: The endpoint is protected by the API Gateway and the Authorization Service, ensuring that only authorized requests can access the data.
  • Empowered Frontend Team: The frontend developers were unblocked immediately and could start developing the mobile app against a real, well-documented API (available in the API Portal).

Mia-Platform Console, with its powerful no-code/low-code components like the CRUD Service, allowed the team to overcome their lack of backend expertise and deliver a critical API on a tight schedule, demonstrating how abstracting common patterns can dramatically accelerate development.